An Optimization Framework for On-Demand Meal Delivery System

Abstract

The success of an on-demand meal delivery business depends on the cost-effectiveness, speedy, and timely delivery of orders. This problem calls for an optimal trade-off between the cost and time of delivery in near real-time for large-scale orders. In this paper, a generic framework for the optimization of first-mile and last-mile of an on-demand meal delivery system is proposed. The objective is to minimize the overall Cost Per Delivery (CPD) and the delay in order delivery time. Few policies are recommended based on the Just-In-Time (JIT) concept and performances are compared within a simulation framework with real order data of a city. The simulation results indicate that the aggressive JIT policies result in substantial savings of CPD by reducing the wait time without compromising the Customer Experience (CX).